Team Member Are :
Shreyas Suvarna - 4so11cs055
Shwetha Patel - 4so11cs056
Vanessa Pinto 4so11cs062
Sumia Shetty - 4so12cs412

Project Guide: Mr Nithin Kumar Heraje, Assistant Professor, Department of Computer Science and Engineering. Abstract: Text Mining refers to the process of deriving high quality information from text. Text mining usually involves the process of structuring the input text, deriving patterns within the structured data and friendly evaluation and interpretation of the output. Minesweep proposes a system that deals with examining the resumes in the most efficient way possible. Minesweep uses the technique of text mining to look for patterns within the resumes which fit the parameters as provided by the organization’s personnel so that the acquired resumes result in the resumes of the best suited candidates as per the job requirements. The existing system is as follows: most large corporations employ a sizeable workforce. At these organizations, job openings at even the bottom of the corporate ladder meet with huge response. The enumerable resumes that are received have to undergo a filtering process. In such situations, the concerning employee(s) is tasked with combing through many resumes to obtain those that best fit the job description by reading each resume with the same careful analysis as the next so as to not to make any errors. When several resumes have been received for any available job posting the charge of having these sorted falls in the hands of a few employees in any organization. This common practice happens to have a huge margin of human error. These said errors include the following: a resume could be mistakenly omitted while leafing through the rest; a person’s concentration is always prone to wither with the monotony of the resumes hence leading the person to skim through them thus possibly missing significant information or also allowing the approval of incorrect data. Along with these given errors, the sorting of resumes is not only tiring but also extremely time-consuming especially in cases of large companies where multiple and varied and several incompetent resumes are received for an open job. Minesweep is an application that can be used to filter numerous resumes as per the criteria that will be given by the user. Minesweep uses the concept and technique of text data mining for the resumes which helps us to filter the resumes based on rules that the user shall input. In large companies where the response to any job availability is received with great enthusiasm by numerous applicants, this application can be put to use to lighten the burden of those tasked with filtering the multiple resumes with the guarantee of no human error that may occur otherwise. The algorithm that has been used extracts the required information from text. It is designed to operate on the resumes, one resume at a time. Here the algorithm is applied on the resumes of the candidates which are considered as the transaction. The algorithm filters each resume character-wise based on rules and the respective existing synonyms that the user will input. After the search is made), the required copies are extracted, and we get the required resumes. Minesweep is a software based on the concept of Text Mining that can be used by the employee/Managers throughout the organization or of a particular department. It helps in reducing the manual work of filtering out the resumes which are prone to human errors. This project is mainly used in the corporate world to filter the resumes and extract them based on the criteria given. Thus by using Minesweep the manager will be able to select the best candidate suited for the organization. Minesweep provides a user-friendly interface which makes the software easy to use. In this fast based and technologically oriented world where time is a major constraint, Minesweep goes a long way in reducing the burden resulting from the manual system.


Team Member Are :
Evangeline Anusha Soans - 4so11cs021
Melisha Maria Menezes - 4so11cs039
Russel Canute Monteiro - 4so11cs050
Sharath Kumar GS - 4so11cs052

Project Guide: Ms Sujatha M - Assistant Professor, Department of Computer Science & Engineering Abstract: FarmBook is a social network service provided through online services. It includes features such as a forum and online shopping. FarmBook is a web-based service that allows farmers to create a public profile, to create a list of users with whom to share connections. FarmBook allows farmers to share ideas, pictures, posts, activities, events, interests with other farmers in their network. It also facilitates them to update prices of commodities so as to enable all farmers to obtain the aforementioned prices. FarmBook also includes a forum which will enable farmers to hold discussions on several of their day to day problems. Their queries can be answered by other farmers or also can be answered by experts in that field. FarmBook also facilitates farmers to buy or sell their products. An account is not needed (anyone can view the products) for the online shopping aspect and to view the market price page of FarmBook. The primary issues concerning farmers involve finding solutions for their farm related problems. There is no online community through which farmers interact. They often have to travel great distances in order to obtain solutions from agricultural research institutes. The solutions obtained are not always satisfactory and are not obtained on time. In addition to this, there is no way for farmers to obtain information about the daily market prices of commodities and products. Hence, they do not get good prices when they sell through middle men and they are often cheated. The general public have no means of buying organic products directly from the farmers. The major application of FarmBook is to provide the farmers with a common platform to communicate, thus enabling them to state their problems and obtain opinions from other farmers around the world. The E-Commerce effectively eliminates the middle man by enabling farmers to advertise any products they wish to sell or view advertisements put up by other farmers. The market price page provides the day to day market rates of farm products which in turn farmers to obtain better returns on their produce. FarmBook can be accessed through mobile app or directly through the website.

Efficient verifiable privacy preserving protocol

Team Member Are :
Deeksha Premchand Rao - 4so11cs013
Harshini Shetty - 4so11cs024
Rainy Mishal D’souza - 4so11cs046
Roslin Thomas P T - 4so12cs408

Project Guide: Ms Sujatha M - Assistant Professor, Department of Computer Science & Engineering Abstract: Cryptography is the study of information hiding and verification. It includes the protocols, algorithms and strategies to securely and consistently prevent or delay unauthorized access to sensitive information and enable verifiability of every component in a communication. When information is transformed from a useful form of understanding to an opaque form of understanding, this is called encryption. When the information is reverted back into a useful form, it is called decryption. Most of the existing works on data aggregation assume that the involved parties will sincerely follow the protocol. The malicious adversary can easily forge his/her input values to achieve incorrect outcomes or simply lie about the computation results to cheat other parities. The previous protocol focuses on data aggregation to an untrusted aggregator without disclosing each other’s data, while preserving privacy. The project proposes an efficient verifiable privacy preserving protocol (evppp), which can resist cheating by incorporating a trusted aggregator.

Contrast enhancement and intensity preservation of gray-level images using multi-objective particle swarm optimization

Team Member Are :
Anisha Corda
Delita Josna D’Souza
Joyline Melita Mendonca
Shahani Natalia Mendonca

Project Guide: Ms Riana Anto, Department of Computer Science and Engineering Abstract: Images will result in dark and low clarity regions due to severe limitations in the illumination condition when first captured. The contrast enhancement of such images is achieved by maximizing the information content carried in the image via a scale factor. The scale factor enhances the contrast of the image by adjusting the lower pixel values to the average value. This enhancement provides better viewing consistence and effectiveness. The image intensity is preserved by applying cost factor on the images. First the image is contrast enhanced by using the histogram equalization method.The fundamental principle of Histogram equalization is to make the histogram of the enhanced image to have approximately uniform distribution so that the dynamic range of the image can be fully exploited. Because histogram equalization is a point process, new intensities will not be introduced into the image. Existing values will be mapped to new values but the actual number of intensities in the resulting image will be equal to or less than the original number of intensities. Image contrast will be enhanced as long as one can make use of the whole available intensity range. Particle Swarm Optimization is then employed to find the optimal constraints in order to maximize the degree of brightness preservation. Since the contrast enhancement and preservation of intensity are contradictory an improved multi-objective particle swarm optimization procedure is proposed to resolve this contradiction, making use of its flexible algorithmic structure. Main features: • Histogram Equalization: It is a method which is used to obtain a uniform histogram i.e. existing values will be mapped to new values but the actual number of intensities in the resulting image will be equal to or less than the original number of intensities. • Multi-Objective Particle Swarm Optimization: Particle swarm optimization (PSO) is a computational method that optimizes a problem by iteratively trying to improve a candidate solution with regard to a given measure of quality.